open 1 - framos · 2017-04-10 · see the “migration to open evision 1.2” pdf document on for...

[email protected]© Copyright 2011 Euresys s.a. Belgium. Euresys® is a registred trademark of Euresys s.a. Belgium.Other product and company names listed are trademarks or trade names of their respective manufacturers.Euresys reserves the right to modify product specifications and price without previous notice.

Free evaluationof the Open eVision

functionalities

Image Analysis Software Tools

New Features

•64-bitlibraries for C++ and .NET development

•OpeneVisionDongle-BasedLicenseson Euresys parallel or USB dongles

•SeamlesscompatibilitywitheVision6.7.1and OpeneVision1.0 C++ and ActiveX APIs

OpenS T U D I O

OpenE V A L

1.2

1.2

1.2

OpenS T U D I O

OpenE V A L

1.2

1.2

1.2

2

General Features•Easytolearnanduse•Robust,flexibleandpowerful•Accurate:sub-pixelmeasurementandcalibration•Thread-safe•64-bitlibrariesforC++and.NETand32-bitlibrariesforC++,.NETorActiveX development•Compatiblewith - Windows® x86 processor architecture - A wide variety of programming languages and development environments

OpeneVisionLibraries

Open to All Image SourcesTheOpeneVisionlibrariesdonotrelyonanyproprietaryhardwaredevicetorun.Theyareabletoprocessimagesavailableinthehostmemory, whatever their origin.The images to be processed may come from a frame grabber, a scanner, a file, or IEEE1394 (Firewire), GigE Vision or USB cameras. Color and monochrome images are supported.

SupportofthepreviousOpeneVisionandeVisionAPIsOpen eVision 1.2 comes with an alternate set of C++ headers and an ActiveX component that allows develop-ingorportingcodeagainsttheolderAPIthatwassuppliedwitheVision6.7.1(andlower)andOpeneVision1.0.See the “Migration to Open eVision 1.2” pdf document on www.euresys.com for detailed information.

Open eVision 1.2 is a rich suite of reliable, powerful and flexible software tools dedicated to image processing and analysis. Open eVision contains a set of 64-bit and 32-bit libraries designed to be integrated into your C++, .NET or ActiveX application.

The general purpose libraries, EasyImage, EasyColor, EasyObject, EasyMatch, EasyFind and EasyGauge cover appli-cations such as image filtering and enhancement, blob analysis, pattern matching, alignment and metrology.

The mark inspection libraries, EasyOCV, EasyOCR, EasyBarCode and EasyMatrixCode, include functions for optical character recognition, character printing inspection and 1D / 2D barcode reading.

IEEE1394

GigEVision

USB2.0

FrameGrabber

General-PurposeLibrariesOpen eVision includes a set of six general-purpose libraries.

EasyImageTM Image processingEasyColorTM ColorcalibrationandtransformationEasyObjectTMBlobanalysisEasyMatchTM Standard pattern matchingEasyFindTM Geometric pattern matchingEasyGaugeTM Sub-pixelmeasurementanddimen-

sion control

MarkInspectionLibrariesThese robust and automatic libraries support a large range of fonts, standards or symbol types.

EasyOCVTM OpticalCharacterVerificationEasyOCRTM Optical Character RecognitionEasyBarCodeTM Bar Code readingEasyMatrixCodeTM Data Matrix 2D code reading

OpenS T U D I O

OpenE V A L

1.2

1.2

1.2

33

Sub-pixelAccuracyandCalibrationAccuracy is of prime importance to most machine vision applications. To meet this requirement, Open eVision’s metrology (EasyGauge), blob analysis (EasyObject), standard patternmatching (EasyMatch) and geometricpatternmatching(EasyFind)functionsallreturnresultswithsub-pixelaccuracy.Tofurtherincreasetheprecisionofthemeasurements,OpeneVisionincludesanextensivesetofcalibrationfunctions.Thesefunctions,afteracalibrationphasethatcanbeperformedfromtheimageofaprecisedotgridorfromalistofpointcoordinates,compensateallmeasurementsforlensdistortion(correctionofpincushion-ing,barreling),non-squarepixels(correctionofthepixelaspectratio)androtationandperspectivedistortions.Theyalsoprovidethecapabilityofworkinginreal-worldmeasurementunits(suchasmils,micronsormillime-ters),andsupporttheconversionofcoordinatesfrompixelcoordinatestoreal-worldcoordinatesandvice-versa.

SavingandLoadingImagesOpen eVision supports loading and saving images from/to the disk. In addition to the existing image formats -BMP,JPEG,TIFF-OpeneVisionnowsupportsJPEG-2000,PNG(asupportslosslesscompression)andSeri-alized(proprietaryimagefileformatobtainedfromtheserializationoftheOpeneVisionimageobjects)formats.On top of this, Open eVision includes an automaticconversionofthefileformatregardless of the pixel format fromanimageonthediscintothebestmatchingOpeneVisionimageobjecttype.

RegionsofInterestandFlexibleMasksTheprocessingspeedofanimagecanbeacceleratedbyfocusingonaspe-cificregionoftheimage(RegionofInterest)avoidinginterferencesfromtheremainderoftheimage.Thenumberofpixelstoconsideristhenreduced.The processing of all Open eVision functions can berestricted to aRegion of Interest (ROI).Open eVisionsupportsnestedrectangularROIs,whichareorganizedin a hierarchical way in each image. ToaddflexibilitytotheshapeoftheROI,OpeneVisionsupports FlexibleMasks for selected functions of the EasyObject and EasyImage libraries. A mask repre-sents a two-class segmentation of pixels which sepa-ratestheassociatedimageindo-careareas(thatmustbeconsidered)anddon’t-careareas(thatshouldnotbeconsidered).Flexiblemaskssupport complexanddis-connected shapes.

ImageandGraphicDisplayOpen eVision also includes functions to help display the result of the image analysis, and provide interactiv-itywith theuser.Forexample,RegionsOf Interest,EasyGaugemeasurementgauges,blobscomputedwithEasyObject,patternslocatedwithEasyMatchandEasyFind,charactersandcodesreadwithEasyOCR,Easy-MatrixCodeandEasyBarCodecaneasilybedisplayedontopoftheimage.

ImprovedExecutionTimeThankstoSSE2Technology

DebuggingandProfilingFunctions: Open eVision supports modern error tracingfunctions,andprovidespreciseexecutiontimemeasurementfunctionsforapplicationprofiling.

Thread-Safety: Open eVision is multi-thread safe. This means that it is designed to support simul-taneousexecutionbymultiplethreadsonthesameCPU.Thisisparticularlysuitableifyourapplicationincludesindependenttasksandallowsthemtobeexecutedsimultaneously.Butthisalsorequiresthatyourapplicationacceptssharingpieceofdatatobeaccessedbyonlyonethreadatanygiventime.Then,youcandesignsuchapplicationssothateachpartiscontrolledbyaseparatethread.Asmanythreadsasrequiredcanbecreated.

SupportedEnvironmentsOpeneVisionare64-bitand32-bitlibrariesthatrequireaprocessorcompatiblewiththex86instructionset,withMMXextensions.IftheSSEorSSE2extensionsarepresent,theyareused,buttheyarenotrequired.

Regions of Interest are organized in a hierarchical way

Flexible Masks

Open eVision

4

Open eVisionTM Accessories

EvaluationThisfreelydownloadableapplicationallowsyoutoevaluatetheOpeneVisionfunctionalities.OpeneVisionEvalbenefitsfromanintuitivegraphicaluserin-terface.AllOpeneVisionfunctionscanbecalledthrougheasy-to-understandmenusgroupedbylibrary.Afterselectingafunction,youcaninteractivelysetitsparametersthroughdialogboxesandtriggertheexecutionofthefunction.A new Getting Started guide documents the Open eVision Eval Graphical user interface.TotakeyourfirststepsinOpeneVision,OpeneVisionEvalcomeswithnewQuickStartTutorials.TheypointupthemainfunctionalitiesofOpeneVisioninadidacticway.TheQuickStartTutorialscomewithalargesetofimagestopracticethelessonsbyyourself.Finally,OpeneVisionEvalallowsyoutoevaluatetheapplicabilityand performance of Open eVision in your particular application with your images.Downloaditfromwww.euresys.comforfree!

Open eVision

OpenS T U D I O

OpenE V A L

1.2

1.2

1.2

55

LearningStructuredDocumentationOpeneVisioncomeswithacomprehensiveandre-structureddocumentationperprogramminginterface(C++,.NETandActiveX).EachofthethreedocumentationsissplitintoaFunctionalGuideandaProgrammingGuide.The Programming Guide includes a comprehensive Reference Guide in addition to the practical Code Snippets foreachoftheconceptsexplainedintheFunctionalGuide.Thedocumentationisavailableintwoformats:com-piledHTMLfiles,whichareconvenienttosearch,aswellasprinter-friendlyPDFfiles.

PracticalandDidacticProjectandApplicationSamplesThe projectsamplesillustrateconciselyhowtousetheOpeneVisionlibrarieswithaparticularIDE.Aprojectsample exists for each of the supported IDEs.The application sampleprogramsillustratethecombineduseofdifferentlibrariesinaspecificapplication.Avarietyofcombinationandapplicationsarerepresented.BotharedownloadablefromtheOpeneVisionDownloadArea.

DevelopmentOpen eVision Studio 1.2 is the development and prototyping tool of Open eVision.It includesthescriptingfunctionthatgeneratesthecode(C++,VisualBasicActiveX)usingeasytounderstandmenusanddialogboxes.Thecodeiscopied and pasted into your application source code in their preferred IDE. A new Getting Started guide documents the Open eVision Studio graphical user interface.TotakeyourfirststepsinOpeneVision,OpeneVisionStudioisdeliveredwithnewQuickStartTutorials.TheypointupthemainfunctionalitiesofOpeneVisionin a didacticway. TheQuickStart Tutorials comewith a large set of variousimagestopracticethelessonsbyyourself.

OpenS T U D I O

OpenE V A L

1.2

1.2

1.2

6

Open eVisionTM Libraries

EasyImageTM

ImageProcessingLibraryMain Features Convolution and morphologyGeometric transformations Image statistics 16-bitaccuracyprocessing

Typical ApplicationsImage enhancement

Image restoration Presence/Absencecheck

EasyImageincludesoperationsusuallyperformedaspre-processingstepstoimprovetheimagequalityandobtainagoodcontrastbetweenthebackgroundandtheobjectstobeinspected.EasyImagesupportsgray-levelandcolorimages.Selectedmorphologyfunctionsarealsooptimizedforbinary(1-bitperpixel)andbi-levelimages.EasyImageincludesnumerousimageprocessingfunc-tions,suchasenhancementandrestorationbylinearornon-linearfiltering,arithmeticandlogicoperations,geometrictransforma-tionsforimageregistration,histogramanalysisforthresholding,projection,…

Refactoring improving the execution time due to SSE2 technology

FlexibleMasks for selected image analysis functions. They provide a powerful way of restricting the processing to freely parts of the image.

InterestPointDetectorsCannyEdgeDetectorThe Canny detector is known as the optimal edge detector. It operates on a gray-scale BW8 image and delivers ablack-and-whiteBW8imagewherepixelshaveonly2possiblevalues,0and255.Pixelscorrespondingtoedges in the source image are set to value 255 while other pixels are set to value 0. The Canny edge detector offersthreeoptimalcharacteristicsfortheimageprocessingapplications:

•Agooddetection:findasmanyedgesintheimageaspossible•Agoodlocalization:thefoundedgesareascloseaspossibletothe“real”edgesintheimage•Aminimalresponse:asingleedgeresponseisacceptedforeachposition,i.e.avoidingmultiplecloseor

intersecting edge responses

HarrisCornerDetectorThe Harris corner detector is popular due to its strong invariance to rotation, illumination variation and image noise. It operates only on a gray-scale BW8 image. The Harris Corner Detector delivers a vector of points of interest.Thefollowingcharacteristicsareavailableforeverypointofinterest:

•Thecornerposition(pixelcoordinateswithsub-pixelaccuracyifenabled)•Thecornernessmeasure•Themagnitudeofthegradientw.r.t.thedifferentiationscaleσd•ThevalueofthegradientalongtheX-axisw.r.t.thedifferentiationscaleσd •ThevalueofthegradientalongtheY-axisw.r.t.thedifferentiationscaleσd

General-Purpose LibrariesOpen eVision

Harris corner detector

Canny edge detector

77

Functions

Intensity scale transformation functions- Gain / Offset change-Normalization-Uniformization- Lookup mapping

Thresholding- Automatic thresholding

Min residue, max entropy, isodata- Manual thresholding

Single threshold (absolute and relative)Double threshold

-Histogram-basedthreshold

Arithmeticandlogicoperations-Arithmeticoperations:

Add, Subtract Multiply, divideCopyInvert, module, shift

-Logicalandbitwiseoperations:AND, OR, XOR, NOT

- Minimum, maximum- Pixel compare-Histogramequalization

Convolution-Pre-definedfiltersfor

Edge detection: Laplacian, Gradient, Prewitt, Sobel, Roberts

Sharpening: Several high-pass filters

Smoothing: Several low-pass including Gaussian filter and uniform filters

-CustomkernelfilteringKernel creation and management functions

Non-linearfiltering- Morphological operators

Erosion, dilationOpening, closingThinning, thickeningTop-hat filtersHit-and-miss transform:

It detects a particular pattern of foreground and background pixels in an image

Morphological distance-Medianfilter

Geometric transformations-Imageregistration(alignment)-Horizontalandverticalmirroring- Translation, scaling and rotation with optional

interpolation-LUT-based(un)warping

Vector operations-Projection-Profile:sampling(linesegment,path,contour)

and analysis

StatisticsMeasurementof:-Area,binarymoments- Weighted moments- Gravity center- Pixel count and pixel statistics- Minimum and maximum gray-level value- Average, variance and standard deviation- Histogram computation and analysis- Image focus

Noisereductionandestimation- Spatial noise reduction

ConvolutionMedian filters

- Temporal noise reductionRecursive averageMoving averageAverage

- Noise estimationRoot-mean-square noiseSignal-to-noise ratio

OperationoninterfacedvideoframesEliminationoftheinterlacedimagesartifactsbyrebuildingorre-aligningfields

Featurepointdetectors- Harris corner detector- Canny edge detector

Other operations- Overlay- Scalar gradient

General-Purpose LibrariesOpen eVision

8

EasyColorincludesasetofoptimizedcolorsystemstransformationfunctionsandcoloranalysisfunctions.Thecolorsystemssup-portedareRGB,XYZ,L*a*b*,L*u*v*,YUV,YIQ,ISH,LSH,VSH,LCHandYSH.EasyColorprovidesefficientmeanstoconvertimagesbetweenthesesystemsandtotransformcolorimagesintograylevelimagesandviceversa.

OperationPrinciplesAlthoughtheRGB(red,green,blue)representationofcolor imagesiswellsuitedforcolorreproduction(it isusedbymonitorsandcameras),manyotherrepresentationshavebeendesignedforvariouspurposes.Moreparticularly, the “Intensity/Saturation/Hue” color systems are well suited for machine vision applications. Easy-Colorsupportsseveralof them.Theyseparate theachromatic (blackandwhite)component (Intensity) fromthechromaticcomponents(SaturationandHue)whichareusedtodescribecolors.Thisallowsamoreintuitiveinterpretation of colors and is very useful to segment colors while eliminating lighting effects. It is thus required, when doing color image processing, to convert the RGB images coming from the camera to another color space, suchasLSH,ISHorYSH.EasyColorprovidesasetofoptimizedcolorspaceconversionfunctions.AlsoincludedinEasyColoraretraditionalcolorimageprocessingfunctions(suchasBayerpatternconversionandcolorbalancecorrection),aswellaspowerfulcoloranalysisfunctions,whichallowtheusertodetectandclassifycolorobjectsanddefects.Forexample,colorimagesegmentationallowsyoutodecomposeacolorim-ageindifferentregionsbyassigningaclasstoeverypixel.ColorimagesegmentationcanbeusedinconjunctionwithEasyObjecttoperformblobanalysisonthesegmentedregions.Itisalsopossibletofilterpixelsbyselectingrangesofvaluesforeachcomponent,forexample,selecting“olivegreen”pixelsbasedontheirhueonly,withaloose discrimination on the intensity and saturation to eliminate surface and lighting effects.

FunctionsColortransformations

-LookupTables(LUTs)-LUTforspecificusage

Colorimetric systems conversion LUT for gain / offset (color)LUT for color calibrationLUT for color balance:

gamma pre-compensation, white balance

Colorimagecomponents- Merging and extracting image components- Pseudo-coloring

Colorclassificationforsegmentation

Special color formats -YUV422decompression- Bayer patterns to RGB

EasyColorTM

Color Image AnalysisLibraryMain Features Fastconversionto11colorspacesColor segmentationColorverification

Typical ApplicationsFoodinspection

PrintingPCB inspection

Open eVision General-Purpose Libraries

99

EasyObjectTM

Main FeaturesImage segmentationObjectlabelingGeometric feature extractionFlexibleMasksHighperformance,especiallyforlargeimagesandimageswithnumerousobjects

Typical ApplicationsSurface inspection

Packaging inspectionObjectlocation

TheEasyObjectlibraryhandlesimagesegmentation,i.e.thedecompositionofimagesintoseparateobjects,alsocalledblobs.Oncetheobjectshavebeenconstructed,theycanbehandledasindependententities.Variousgeometricparametersorfeatures,suchasarea,width,orellipseofinertia,canbecomputedonthem.Thentheobjectsofinterestcanbeselectedbymeansoftheirpositionoroftheircomputedfeatures.EasyObjectalsosupportstheinspectionofholesindefinedobjects.Holesaremanagedastheobjectsthemselves,benefitingfromthesamegeometricalfeatures.TheEasyObjectlibrarymanagestherelationshipbetweenobjectsandholes,definingparentobjectsforholes.

FlexibleMasksEasyObjectsupportstherestrictionoftheblobanalysisto complex- or disconnected-shape regions of the image thankstotheFlexibleMasksthatareavailableforencodingfunctions.EasyImagecanalsogenerateFlexibleMasksfroman encoded image.

Refactoringgloballyimprovingtheexecutiontime, especially for large images and images with numerous objects.

Object-OrientedAPI EasyObjectisaccessiblethroughabrand-newobjectorientedAPIwhichbringsseveralmajorsimprovementstothelibrary:- Separate classes according to the different concepts such asencoding,objectfeatureextraction,objectselection

-Improvedaccesstothefeaturesoftheobjects.Theobjectsandtheholescanbeefficientlyaccessedrandomly(i.e.inanindex-basedfashion)

Functions Imageencoding

- Construction of the runsSegmentation

Grayscale single thresholdGrayscale double thresholdColor single thresholdColor range thresholdReference imageImage rangeLabeled imageBinary image

Pixel aggregation-Objectconstruction:aggregationoftherunsintoobjects-Holeconstruction:aggregationoftherunsintoholes-Continuousmodeforwebinspectionapplicationsusing

line-scan cameras

Objectfeatureextraction: geometric parameters computation

Objectselectionandsorting according to any feature value

Open eVision

Increased performances for images with numerous objects

General-Purpose Libraries

BlobAnalysisLibrary

10


BlobFeaturesComputableFeatures

Position-Limit(top,bottom,left,right)-Gravitycenter(XandY)-Weightgravitycenter(XandY)

Extents-Area(pixelcount)-Feretbox(centerXandY,height,widthwithdistinctorientationanglesat22,45,68degrees)andBoundingbox(centerXandY,height,width)

-Minimumenclosingrectangle(angle,centerXandY,heath,width)

Miscellaneous-Startingpointoftheobjectcontour(XandY)- Largest run - Run count -Objectnumber(index)

Statistics-Pixelgray-levelvalue(average,deviation,variance) -Pixelgray-levelvalue(minandmax)

EllipseofInertia- Eccentricity of the ellipse of inertia- Ellipse- Second order geometric moments

ConvexHull

Theconvexhullofashapeistheconvexpolygonofminimumareathatcompletelysurroundsanobject.Theconvexhullcanbeusedtocharacterizetheobjectfootprint,aswellastoobserveconcavities.

GraphicRepresentation

Theobjectscanbedrawnontothesourceimage. Thefollowingblobfeatureshaveapre-setgraphicalrepresentation:-Objects- Diagonals-Boundingbox- Convex hull- Ellipse-FeretBox-Feretboxwithanangleof22°-Feretboxwithanangleof45°-Feretboxwithanangleof68°- Gravity center- Minimum enclosing rectangle- Weighted gravity center

1111

EasyMatchTM

Main Features NormalizedcorrelationmethodSub-pixelaccuracyRotation and scaling supportMultiple pattern occurrences Gray-level and color images supportNon-square pixels management Don’t care areas

Typical ApplicationsPresence,absence

AlignmentPick and place

EasyMatchisagray-levelandcolorpatternmatchinglibrary.Itletsyoutrainthesystemonareferencepatternandafterwardslocateitsoccurrencesinotherimages.Thistoolisveryconvenientwhenthepositionofagivenpartisunknowninthefieldofview,orifthepresenceofpartsmustbecontrolled.Thelibraryworksbyusingnormalizedcorrelationmethod,i.e.measuringdiscrepanciesbetweenthepatternandthetargetimage.

FeaturesMultiple pattern occurrencesSeveraloccurrencesofapattern,uptoauser-definednumber,arereturned.Onlythereliableonesare retained.

Standard,offset-normalized,gain-normalizedandfullynormalizedcorrelationThecorrelationiscomputedoncontinuoustonevalues(asopposedtobinary).Itiswellknownthatwhen the lighting conditions vary, as it is often the case, straight comparison of the pattern and imagebehavesbadly.Tocopewiththis,automaticallyadjustingthecontrastand/orintensityofthepatternbeforecomparisonisveryeffective.Thisprocessisknownasnormalization.EasyMatchprovidesfourdistinctnormalizationmodes,dependingonwhetheragainand/oroffsetcompensa-tion is used.

Normal,inverseormixedcontrastBecauseofparticularlightingeffects,anobjectcanappearwithinvertedcontrast(whiteonblackinsteadofblackonwhiteorconversely).Dependingontheapplication,itcanbeusefultokeepinvertedinstancesortodisregardthem.Threecontrastmodesareavailable:considerpositiveoc-currencesonly,negativeoccurrencesonlyorboth.

Translation,rotationandisotropic/anisotropicscalingTofindthebestmatchesbetweenthepatternandtargetimage,thetargetisallowedtotranslatehorizontallyandvertically.Ad-ditionally,itcanbeallowedtorotateand/ortochangeitsscaleintheXandYdirectionssimultaneouslyorindependently.Therotationangleandscalefactorsvaryinauser-specifiedinter-val.Alldegreesoffreedomcanbecombinedatwill.

Variableaccuracy,uptosub-pixellevelTheaccuracywithwhichthepatternismeasuredcanbecho-sen (the lessaccurate, the faster).Aone tenth-of-a-pixelac-curacycanbeachieved.

Don’tcarepixelsWhenthepatterncannotbeinscribedinarectangularROI,thesurroundingofthepatterncanbeignoredbysettingthepixelsvaluesbelowathresholdlevel.Thesepixelswillnottakepartinthematchingprocess.Thesamefeaturecanbeusedifpartsofthetemplatechangefromsample to sample.

Gray-scaleandcolorimagesEasyMatchprocesses8-bitgray-scaleimagesaswellas24-bitRGBimages.

Non-squarepixelsWhenimagesareacquiredwithnon-squarepixels,rotatedobjectsappearskewed.Takingthepixel aspect ratio into account can compensate for this effect.

PatternMatchingLibrary


12

EasyFindTM

Main Features FeaturepointtechnologyFullyautomatic,fastandrobustRotation and scaling invariantHigh tolerance to pattern degradation Don’t care areas User-definedpivotpoint

Typical ApplicationsPresence,absence

AlignmentPick and place

Printing industry

Basedonaninnovativefeature-pointtechnology,EasyFindisdesignedtorapidlyfindoneormoreinstancesofareferencemodelintheimage.Comparedtoclassicalalgorithms,EasyFindfeaturesfasterprocessingandimprovedrobustness.Euresys’geomet-ricpatternfindershowsexcellentperformanceswhenhandlinginstancesthatarehighlydegradedduetonoise,blur,occlusion,missingpartsorunstableilluminationconditions.Withanadjustableaccuracyuptosub-pixellevel,EasyFindreportsverypreciseinformationabouttheinstancesfound,suchastheirlocation,rotationangle,scaleandmatchingscore.EasyFindsupportsdon’tcareareas.Thisfeatureallowsthecreationofcomplexpatternshapes.

FastProcessingandImprovedRobustnessEasyFindisbasedonanovelfeature-pointtechnology.Insteadofcom-paring the reference model to the sample image pixel-wise, it carefully selectssalient features in themodel.ThismethodallowsEasyFind tomatchonlytheareasthatconveyvaluableinformation,resultinginfasterprocessingandmuchimprovedrobustness.EasyFindalso bringsa newapproach to the problemof detecting in-sufficientlydefinedmodels inan imagebyproposing threeselectablepatterntypes:Consistentedges

Thismode is used for patternswithwell defined edges or to findnon deformed instances. It detects instances highly degraded due to blur,noise,occlusionorvaryingilluminationconditions.A pointbypointscoringmethod makes this operating mode more resilient to large occlusions and/or largevariationsofcontrast.Italsogloballyreducesthecomputationtimeofthefindingphase.

Thin structureThismodeisusedto locatepatternsincludingparticularlythinstructures.It isrobustagainstblur,noise,occlusion and illumination variation.

ContrastingregionsThismode isused forpatternswithpoorlydefinededgesorpatternsexhibitingnoise,blur,and randomtexture.Thecontrastingregionspatterntypeisrobustagainstblur,noise,illuminationvariation. These three patterntypesallowEasyFindtomaximizetheaccuracyandthespeedoftheprocessing,evenindifficultcases.

ThesethreepatterntypesallowEasyFindtomaximizetheaccuracyandthespeedoftheprocessing,evenindifficultcases.

GeometricPatternMatchingLibrary


1313

EasyGaugeTM

Sub-pixelMeasurement andDimensionControlLibraryMain Features Sub-pixelpointlocationandedgefitting HighlyaccurateandrobustPosition,orientation,size,curvature,distances AdvancedandautomaticcalibrationMultiple gauge models Graphical model edition

Typical ApplicationsGauging applicationsCalibrationmetrologyAssemblyinspection

EasyGaugeisacutting-edgemeasurementanddimensioncontrollibraryforuseingaugingandmetrologyapplications.Byrely-ingonprovensub-pixeledgedetectionandshapefittingalgorithms,itallowsdeterminingthedimension,position,curvature,size,angleordiameterofmanufacturedpartswithanexcellentaccuracy.Robustnessisensuredbypowerfuledge-pointselectionmech-anismsthatareintuitiveandeasytotune,allowingmeasurementinclutteredimages.Inadditiontothesestate-of-the-artfeatures,EasyGaugealsosupportstheautomaticmeasurementofparallelsides,thusprovidingmeansofmeasuringthethicknessofflatorbentobjects,aswellasthepreciselocationofcorners.

AdvancedandAutomatic CalibrationEasyGauge has advanced built-in calibra-tion capabilities to transparently convertpixel measurements to physical units; this relieves the user of the need to convert co-ordinates. Non-square pixels and rotated coordinate axis are supported. EasyGauge also provides means to determine and cor-rect perspective and optical distortion, with no performance loss.

Gauge Grouping EasyGauge supports grouping of the measurement gauges and lets these groups track the measured items intheimage.Thesecanfreelytranslateand/orrotatewhiletheprobesarerepositionedaccordingly.Derivedmeasurementssuchasdistancesbetweenfeaturepointscanthenbecomputed.


14

EasyOCVTM

OpticalCharacterVerificationLibraryMain Features Comprehensive automatic training Gray-scale analysis Text and character-level inspection

- Contrast, position, shape defect detection - Allowed text translation, rotation, character translation … - Statistical training - …

Typical ApplicationsMark inspection Labelinspection

Lotmixingverification

EasyOCVisanopticalcharacterverificationtoolusedformarkinspection.Itprovidesautomatictrainingofthemodel,adjustableacceptancelevelsandarobustgray-scaleanalysis.EasyOCVissuitableforvariousmarkinspectionandlabelprintingverificationapplications.EasyOCVhasbeendesignedforon-lineverification.Itcandetectblurred,misaligned,distortedordoublemarks,extraormissingink,missingcharactersorreversedmarksaswellascontrastproblems.

WorkingProcessEasyOCVfirst requires trainingontheparticularmarking.During thisphase,agoodquality template ispre-sentedtothesystem.Aninteractiveutilityallowsdefiningthestructureofthetemplateandrelatedacceptancecriteria,indetails.Thetemplatelearntcanbesavedforlateruse.When a template model is ready, the inspection may take place. The sample image is processed and the system firstlocatesthemarking,allowingittobetranslated,rotatedandre-scaledorevenshearedwithrespecttothetemplate. After location, geometric comparisons are performed and a series of matching scores, called quality indicators, are computed. When these fall outside of given acceptance intervals, a defect warning is reported.

Mark Inspection LibrariesOpen eVision

1515

EasyOCRTM

CharacterRecognitionLibraryMain Features TeachablesystemReliableandrobustrecognitionSizeinvarianceTrained character fontsBroken character reconstruction Touching character separation Pre-definedfonts

Typical ApplicationsPartidentificationParttraceability

Serialnumberverification

EasyOCR isa font-dependentprintedcharacter reader,basedona templatematchingalgorithm. Itsupportsgray level images.EasyOCRallowstrainingonthefonttoberecognizedbyshowingsampleimagesofallpossiblecharacters.Forthisreason,itisabletoreadanykindofshorttext(serialnumbers,labels,...)suchasthosefoundinindustrialenvironments.

WorkingProcessThetrainingphaseusesaninteractiveutilitytoshowsamplesofthecharactersandstoretheminafontfile.Moreover,EasyOCRprovidesthemeans for you to write your own, custom, font editor.EasyOCRalsoprovidesthreestandardfonts:OCR-A,OCR-BandSemi.Thisavoidstocollectsampleswhenhavingtorecognizeatextusingone of these fonts. EasyOCRusesblobanalysisfunctionstosegmenttheimageandextractthecharactersconstitutingthetexttoberead.Blobsareselectedascharactersbasedontunablesizeandshapecriteria.Moreover,EasyOCRisabletodealwithcharacterssplitintoseveralblobs.Whentheexactposition of the characters in the image is unknown, EasyOCR functions will process the entire image and locate the characters.


1616

EasyBarCodeTM

BarCodeReadingLibraryMain Features AutomaticbarcodedetectionVeryfastandrobustFullsupportofnumeroussymbologies

Typical ApplicationsProductidentificationBarcodeverification

Parttraceability

EasyBarCodeisalibrarydesignedtoautomaticallylocateandreadbarcodes.Barcodesencodeshortcharacterstringandarewidelyusedformarkingandidentifyinggoods.EasyBarCodeisabletoidentifyandreadawiderangeofstandardcommonly-usedsymbologiesaswellasspecialsymbologies.EasyBarCodeautomaticallylocatesthebarcodesymbolintheimageandsupportscoderotation.Moreover,forprototypingorspecialcases,anadvancedmanuallocationmodeisalsoavailable.

SupportedSymbologiesStandardsymbologies

Codabar Code39 MSI

Code 128 EAN 128 UPC A

Code 25 Interleaved EAN13 UPC E

Additionalsymbologies

Binary Code Code32 EAN 8

Code ABC Anker Code39Extended,Reduced IBM Delta Distance A

CodeBC412 Code93 Plessey

Code 11 Code93Extended Telepen

Code 25DataLogic, Matrix, IATA, Industry, Compressed, Inverted

Code BCD Matrix

Code CIP

Code STK


1717

EasyMatrixCodeTM

2DDataMatrixCodeReadingLibraryMain FeaturesImpressiverobustnesstonoise,bluranddistortionAutomatic code detection Very fast operation Error detection and correction RotationandflippinginvariantScalinguptoaminimumsizeMinimumcell3X3pixelsContrast invariant Supported codes Data Matrix codes, including ECC200, ECC000, ECC050, ECC080,ECC100andECC140encodingtypes Automatic compensation for illumination changes

Typical ApplicationsParttraceability

Serialnumberverification

DataMatrixcodesarewidelyusedforparceltrackingandpartidentificationinthesemiconductor,pharmaceuticalandmechanicalindustries.EasyMatrixCodeisafullyautomaticreaderof2DDataMatrixcodes.Itrecognizessymbolsofanysize,contrast,locationandorientationinasingleoperation.EasyMatrixCodesupportsgraylevelimages.Asanautomaticlibrary,alearningphaseisnotrequiredbutisanoption.Inaddition,thecharacteristicsofthematrixcodemaybesetmanuallyallowingtorestrictthesearchandthustoacceleratetheprocess.Thecharacteristicsareforexample:familyofthematrixcode,numberofcells,blackcellsonwhitebackgroundorreverse,flippingallowedornot.Errordetectionandcorrectionalgorithmsareusedtoprovideareliablereading.EasyMatrixCodeisfullycompatiblewiththeANSI/AIMBC11-1997standard.

ImpressiveRobustnesstoNoise,BlurandDistortionEasyMatrixCodecanbeusedinthemostdemandingsituation.Ithasbeendesignedtofulfillthemostdifficultapplicationcasessuchas:Badilluminationconditions

- Non uniform contrast- Under or over exposures

Badimagesresultingfromprintingoropticaldefects- Blurred Data Matrix codes- Anisotropic and non uniform scaling- Noisy images- Skewed images

Codeshardtodetect-Smallsizematrixcodes-Matrixcodesonatexturedbackground- Laser marked- Nailed- Rectangular matrix codes

PrintQualityVerification EasyMatrixCode inspects the quality of the printed matrix code. Indicators of the printing quality are defined and graded by theANSI/AIM standard -InternationalSymbologySpecification;DataMatrix-.


Open eVision - Libraries, October 2011 DS_OpeeVisLib_Oct11_Cor0

www.euresys.com [email protected]

Your distributorAmerica, Euresys Inc. 27126-B Paseo Espada, Suite 704 - San Juan Capistrano, CA 92675, U.S.A. Toll free: 1.866.387.3797 - Phone: 1.949.743.0612 - Fax: 1.949.743.2790Asia, Euresys Pte. Ltd. 750A Chai Chee Road, #07-15 Technopark @ Chai Chee, Singapore 469001 Phone: +65 6445 4800 - Fax: +65 6445 8849

Japan, [email protected]

Europe, Euresys s.a., Corporate Headquarters 14, Avenue du Pré-Aily, B-4031 Angleur, Belgium Phone: +32 4 367 72 88 - Fax: +32 4 367 74 66

Free evaluationof the Open eVision

functionalities

OpenS T U D I O

OpenE V A L

1.2

1.2

1.2

Open eVision Ordering Information

Open eVision for USB Dongle Libraries

Product Code Designation4151 Open EasyImage 1.2 for USB Dongle4152 OpenEasyObject1.2forUSBDongle4153 Open EasyMatch 1.2 for USB Dongle4154 Open EasyColor 1.2 for USB Dongle4155 Open EasyOCR 1.2 for USB Dongle4156 Open EasyOCV 1.2 for USB Dongle4157 Open EasyMatrixCode 1.2 for USB Dongle4158 OpenEasyFind1.2forUSBDongle4159 Open EasyGauge 1.2 for USB Dongle4160 Open EasyBarCode 1.2 for USB Dongle

Bundles and SDKProduct Code Designation

4164 OpenInspectionbundle1.2forUSBDongle4165 Open Mark Inspection Bundle 1.2 for USB Dongle4167 OpenFullBundle1.2forUSBDongle4172 Open eVision SDK 1.2 for USB Dongle

Open eVision StudioProduct Code Designation

4163 eVision/Open eVision USB Dongle DongleProduct Code Designation

6512 eVision/OpeneVisionUSBDongle(empty)

Open eVision for Parallel Dongle Libraries

Product Code Designation4201 Open EasyImage 1.2 for PAR Dongle4202 OpenEasyObject1.2forPARDongle4203 Open EasyMatch 1.2 for PAR Dongle4204 Open EasyColor 1.2 for PAR Dongle4205 Open EasyOCR 1.2 for PAR Dongle4206 Open EasyOCV 1.2 for PAR Dongle4207 Open EasyMatrixCode 1.2 for PAR Dongle4208 OpenEasyFind1.2forPARDongle4209 Open EasyGauge 1.2 for PAR Dongle4210 Open EasyBarCode 1.2 for PAR Dongle


4214 OpenInspectionbundle1.2forPARDongle4215 Open Mark Inspection Bundle 1.2 for PAR Dongle4217 OpenFullBundle1.2forPARDongle4222 Open eVision SDK 1.2 for PAR Dongle


4213 eVision/Open eVision Parallel Dongle DongleProduct Code Designation

6513 eVision/OpeneVisionParallelDongle(empty)

Libraries

Product Code Designation4251 Open EasyImage 1.2 for Soft-Based Licensing4252 OpenEasyObject1.2forSoft-BasedLicensing4253 Open EasyMatch 1.2 for Soft-Based Licensing4254 Open EasyColor 1.2 for Soft-Based Licensing4255 Open EasyOCR 1.2 for Soft-Based Licensing4256 Open EasyOCV 1.2 forSoft-Based Licensing4257 Open EasyMatrixCode 1.2 for Soft-Based Licensing4258 OpenEasyFind1.2forSoft-BasedLicensing4259 Open EasyGauge 1.2 for Soft-Based Licensing4250 Open EasyBarCode 1.2 for Soft-Based Licensing


4264 OpenInspectionbundle1.2forSoft-BasedLicensing4265 Open Mark Inspection Bundle 1.2 for Soft-Based Licensing4267 OpenFullBundle1.2forSoft-BasedLicensing


4263 Open eVision Studio 1.2 for Soft-Based Licensing

OpeneVisionDongle-BasedLicensesDongle-BasedLicensesoffertheflexibilitytobetransferredfromaPCtoanother.TopurchaseaDongle-BasedLicense,selectoneoftheEuresysdongles(USBorParallel)plusthedesiredlicense(s)tobestoredonthisdongle.

OpeneVisionSoftware-BasedLicensesSoftware-basedlicensesdonotrequireanydongle,theyarelinkedtothePConwhichtheyhavebeenactivated.

AnOpeneVisioncustomerisfreetochooseamongalargechoiceofproductsthemostsuitableandattractiveofferforhis application.Alllibrariescanbepurchasedindividually.OpeneVisionBundlesgroupseverallibraries:- The Open eVision Inspection bundleincludesEasyImage,EasyColor,EasyObject,EasyMatchandEasyGauge.- The Open eVision Mark Inspection bundleincludeEasyOCR,EasyOCV,EasyBarCodeandEasyMatrixCode.The Open eVision SDK group the following Open eVision products: Open eVision Studio, EasyImage, EasyColor, EasyObject, EasyMatch,EasyFind,EasyGauge,EasyOCR,EasyOCV,EasyBarCodeandEasyMatrixCode.

open 1 - framos · 2017-04-10 · see the “migration to open evision 1.2” pdf document on for...

Documents